The long-term objective of this project is the noninvasive in situ characterization of regional concentrations, fluxes and mobilities of Magnetic-Resonance (MR)-detectable metabolites in brain for several disorders. For reasons of MR sensitivity we will use proton spectroscopy, which can detect metabolites such as N-acetylaspartate (NAA), cholines, creatines, lactate, inositols, glutamate, glutamine, adenosines, gamma-- aminobutyric acid (GABA), glycine and taurine. Metabolite concentrations, fluxes and mobilities sensitively reflect cell function (e.g. biochemistry, compartmentation, bonding, exchange, viscosity) and are expected to respond quickly to cellular changes. They may therefore provide a means to rapidly diagnose and evaluate brain disorders in situations where other techniques fail. Our proposed research focusses on evaluation of brain viability after ischemic injury. To assess the extent of injury we will compare metabolite intensities, fluxes and mobilities in healthy and injured brain locations. Ischemic regions will be outlined by water diffusion imaging. Intensity measurements will focus on NAA to assess neuronal viability (see hypothesis 1), on GABA to study long-term ischemic damage (see hypothesis 2) and on lactate to study its distribution with respect to the ischemic area. We will mainly assess concentration ratios, but preliminary work on the determination of absolute concentrations will also be performed. Correlation between changes in the anisotropic apparent diffusion constant of NAA and the apparent water diffusion constant will be used to assess their relation with respect to cytotoxic edema in ischemic injury (see hypothesis 3). To address tissue viability we will study the rate of 13C-glucose turnover as measured by ongoing lactate production in ischemic regions (see hypothesis 4) and by turnover into glutamate via the tricarboxylic acid cycle in healthy tissue (see hypothesis 5). Comparison of the region of active lactate turnover with regions of high equilibrium lactate concentration will be used to asses necrosis. In order to test these hypotheses, the first aim is to design quantitative MR-methods for monitoring of metabolite intensities and fluxes in certain locations (localized spectroscopy) or as a function of brain anatomy (spectroscopic imaging: SI). Using proton MRSI, a spatial resolution comparable to positron emission tomography (PET) should be attainable. Editing techniques will be designed that can selectively and/or simultaneously detect the mentioned metabolites and fluxes of enriched metabolites. The second aim is to determine metabolite intensities and fluxes for healthy and injured tissue in animals and humans. In the animal model (middle cerebral artery occlusion in cat), NAA mobility will also be; measured and changes in the mentioned quantities and in the location of the injury will be studied as a function of time after the injury. A major effort will be on automation of these functional imaging methods for use in the clinic.